Abstract While other leading causes of death have seen significant declines, suicide rates in youth have increased. It is the second leading cause of death among adolescents, and hospital encounters for suicide in this age group have doubled over the last decade. In 50 years of research, the field has focused almost exclusively on long- term risk for suicidal outcomes, with an average follow-up period of 10 years. Indeed, there has been only one study of short-term risk for suicidal behavior in adults and none in adolescents. This is notable because long- term risk factors tend to be poor predictors of short-term risk. Consequently, the current evidence base for healthcare providers' decisions in determining proximal risk and the corresponding appropriate clinical response is quite limited. The low base rate of suicidal behavior in adolescents is particularly challenging for studying short-term risk for this outcome, and this in large part accounts for the absence of research in this area. To overcome this difficulty, large samples drawn from high-risk populations are necessary. In the current proposal, we are uniquely positioned to address this difficulty. That is, we will leverage secondary data from one of our current studies of suicide risk in adolescent psychiatric inpatients by pooling these data with data collected on 200 new patients for a final sample of 400. Additionally, recent developments in ambulatory assessment technologies offer novel opportunities for elucidating the temporal dynamics of proximal suicidal risk, thereby addressing the elusive question of not only whether an individual is at risk for engaging in suicidal behavior in the near future but when they may be most at risk. The current application will evaluate arousal (in the form of stress and sleep disturbance) in relation to proximal risk for suicidal outcomes using ambulatory assessment measures. Specifically, employing ecological momentary assessment, a mobile biosensor, and actigraphy, we aim to evaluate the interrelation of time-varying stress (at psychosocial and physiological levels) and sleep irregularity in the temporal dynamics of risk for suicidal ideation and events (suicide attempts and intensive clinical care for acute suicidality) over one month post-discharge. Given the prefrontal cortex's involvement in regulating arousal, we will also evaluate whether executive functioning at baseline moderates the effects of time-varying stress and sleep irregularity on proximal risk for these suicidal outcomes. Finally, we will supplement traditional statistical methods with novel advances in computational psychiatry. This proposal is innovative by being uniquely powered statistically to clarify the temporal dynamics of proximal risk for suicidal events, and leveraging new technologies to combine multiple methods (i.e., behavioral, physiological, neurocognitive, and self-report measures), including three continuous streams of ambulatory assessment data, across multiple units of analysis (i.e., psychosocial and physiological stress). This application may identify important variables for monitoring of patients during periods of elevated risk and identify specific and tractable treatment targets, thus providing directions for future clinical intervention efforts.